Power transmission line thunder and lightning monitoring method, device and equipment and readable storage medium

Abstract

The invention provides a power transmission line lightning monitoring method, device and equipment and a readable storage medium, and the method comprises the steps: monitoring the current of a power transmission line to determine a target moment, the current of the power transmission line being greater than or equal to a first preset threshold at the target moment, and being less than the first preset threshold at the previous moment of the target moment; acquiring a monitoring image of the power transmission line in a period of time before and after the target moment as a to-be-detected image; and performing image detection on the to-be-detected image to obtain a judgment result of the lightning event. According to the application, the current influence of thunder and lightning on the power transmission line is utilized, and most of redundant data without a monitoring target is eliminated from the monitoring image, so that the image processing amount is reduced, the equipment computing power resource of the system is saved, and the system cost is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of power transmission lines, and in particular to a method, device, equipment and readable storage medium for lightning monitoring of power transmission lines. Background Art

[0002] In recent years, with the advancement of industrialization and rapid economic development, the construction of power facilities in my country is in full swing. The increase in the scale of power transmission has also increased the difficulty and complexity of power transmission line security. In tracing the root cause of power transmission line failures, natural disasters are often an important inducing factor and prevention target. Among the many natural disasters, the proportion of disasters caused by lightning is particularly prominent, accounting for about 60%-80% of natural disasters in transmission lines. Therefore, monitoring and locating lightning events on transmission lines can greatly improve the ability of power transmission safety and fault tracing and maintenance.

[0003] In the related technology, lightning events on transmission lines are monitored and located based on image detection, meeting the complex and diverse needs of monitoring targets, and presenting the data analysis results intuitively and visually. At the same time, the background information contained in the image data can also be used for more data analysis, such as environmental assessment, disaster assessment, etc., making the power monitoring system comprehensive and information-based, and better helping maintenance personnel to trace faults and repair damage. However, in the current transmission line lightning monitoring system based on image detection, real-time image detection is performed on the monitoring images of the transmission lines, and there is a large amount of redundant data without monitoring targets in the monitoring images. Image monitoring of these redundant data will ineffectively consume the system's equipment computing resources and increase system costs. Summary of the invention

[0004] The present application provides a method, device, equipment and readable storage medium for lightning monitoring of power transmission lines, which can solve the technical problem in the prior art that redundant data ineffectively consumes the computing power resources of the system equipment.

[0005] In a first aspect, an embodiment of the present application provides a method for monitoring lightning in a power transmission line, the method comprising:

[0006] Monitoring the current of the transmission line to determine a target time, wherein the current of the transmission line is greater than or equal to a first preset threshold at the target time and is less than the first preset threshold at a time before the target time;

[0007] Acquire monitoring images of the power transmission line within a period of time before and after the target time as images to be tested;

[0008] Image detection is performed on the image to be tested to obtain the determination result of the lightning event.

[0009] Furthermore, in one embodiment, the step of performing image detection on the image to be detected to obtain a determination result of a lightning event includes:

[0010] Pre-process the image to be tested to highlight the contour features of lightning;

[0011] The preprocessed image to be tested is input into the trained target detection model to obtain the target detection result, wherein the target detection model is used to detect lightning objects and power transmission line objects;

[0012] If there are lightning objects and transmission line objects in the target detection results, and the positional relationship between the lightning objects and the transmission line objects meets the preset requirements, the determination result of the lightning event is determined as that there is a risk of lightning striking the transmission line.

[0013] Furthermore, in one embodiment, the preset requirements include:

[0014] The center of the detection frame of the lightning object is located in the target area, the center of the target area coincides with the center of the detection frame of the transmission line object, and the ratio of the area of ​​the target area to the area of ​​the detection frame of the transmission line object is a second preset threshold, and the second preset threshold is less than 1.

[0015] Furthermore, in one embodiment, the preset requirements include:

[0016] The intersection-and-joint ratio of the detection frame of the lightning object and the detection frame of the power transmission line object is greater than a third preset threshold.

[0017] Furthermore, in one embodiment, the step of preprocessing the image to be detected to highlight the contour features of lightning includes:

[0018] Convert the image to be tested from RGB color space to Lab color space;

[0019] In Lab color space, contrast-limited adaptive histogram equalization is performed on the brightness channel L;

[0020] Convert the processed image to be tested from Lab color space to RGB color space.

[0021] Furthermore, in one embodiment, the training set used by the target detection model includes original training data and expanded training data generated by the original training data through image data enhancement operations, and the image data enhancement operations include color transformation, scaling, and cropping and splicing;

[0022] The expanded training data generated by color transformation is used to reduce the background interference on the detection of lightning objects and power transmission line objects;

[0023] The expanded training data generated by scaling is used to enhance the adaptability of the target detection model to lightning objects and power transmission line objects of different scales;

[0024] The expanded training data generated by cropping and splicing is used to enhance the adaptability of the target detection model to lightning objects and power transmission line objects under different background, weather and lighting conditions.

[0025] Further, in one embodiment, the current of the transmission line is synchronously sampled with the monitoring image of the transmission line, and the sampling frequency is greater than or equal to 10 Hz;

[0026] The image to be tested is the monitoring image of the transmission line within 0.1s before and after the target time.

[0027] In a second aspect, an embodiment of the present application further provides a transmission line lightning monitoring device, the transmission line lightning monitoring device comprising:

[0028] A current detection module, used to monitor the current of the transmission line to determine the target time, wherein the current of the transmission line is greater than or equal to a first preset threshold at the target time and is less than the first preset threshold at a time before the target time;

[0029] An image screening module is used to obtain monitoring images of the power transmission line within a period of time before and after the target time as images to be tested;

[0030] The image detection module is used to perform image detection on the image to be detected to obtain the determination result of the lightning event.

[0031] In a third aspect, an embodiment of the present application further provides a transmission line lightning monitoring device, which comprises a processor, a memory, and a transmission line lightning monitoring program stored in the memory and executable by the processor, wherein when the transmission line lightning monitoring program is executed by the processor, the steps of the above-mentioned transmission line lightning monitoring method are implemented.

[0032] In a fourth aspect, an embodiment of the present application further provides a readable storage medium, on which a transmission line lightning monitoring program is stored, wherein when the transmission line lightning monitoring program is executed by a processor, the steps of the above-mentioned transmission line lightning monitoring method are implemented.

[0033] In the present application, the current of the transmission line is monitored to determine the target time, wherein the current of the transmission line is greater than or equal to a first preset threshold at the target time, and is less than the first preset threshold at the moment before the target time; a monitoring image of the transmission line within a period of time before and after the target time is obtained as the image to be tested; and image detection is performed on the image to be tested to obtain the determination result of the lightning event. Through the present application, the current impact caused by lightning on the transmission line is utilized to eliminate most of the redundant data without monitoring targets from the monitoring image, thereby reducing the image processing amount, saving the system's equipment computing resources, and reducing system costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of a process flow of a method for monitoring lightning in a power transmission line according to an embodiment of the present application;

[0035] Figure 2 for Figure 1 Detailed flow diagram of step S3 in FIG.

[0036] Figure 3 for Figure 2 Detailed flow chart of step S31;

[0037] Figure 4 This is a schematic diagram of the functional modules of a lightning monitoring device for a power transmission line in one embodiment of the present application;

[0038] Figure 5 This is a schematic diagram of the hardware structure of the power transmission line lightning monitoring device involved in the embodiment of the present application. DETAILED DESCRIPTION

[0039] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0040] As for lightning location technology around power transmission lines, the current mainstream detection methods are mostly based on the sound, electromagnetic field and light characteristics of lightning. Among them, the lightning whistle lightning detection system uses the huge sound signal before the lightning appears as the detection signal of lightning, and then infers the timing of lightning through the time delay algorithm. However, this method relies heavily on the accuracy of the time delay algorithm to estimate the timing and location of lightning, which is limited by terrain factors. Complex environment and terrain may cause signal attenuation and reflection, greatly reducing the effective detection range of the system and the accuracy of lightning spatiotemporal information prediction.

[0041] Among the location detection methods that use the electromagnetic field characteristics of lightning, the most representative one is the lightning network detection technology, which is a lightning monitoring technology based on very high frequency (VHF). This method uses the strong electromagnetic radiation generated during the discharge of lightning, covering multiple frequency bands from low frequency to high frequency and very high frequency (VHF). High-frequency electromagnetic waves can not only penetrate the cloud layer, but also have a long propagation distance and are less susceptible to interference. Through the network coordination domain composed of multiple receiving stations, a specific VHF frequency band is monitored, and with the assistance of time synchronization equipment, the spatiotemporal position of lightning discharge and its development process are accurately calculated. Although this mainstream lightning monitoring technology has the characteristics of high real-time performance and a wide monitoring range, it still has some shortcomings. Since VHF monitoring of lightning requires multiple stations to work together, it indirectly increases the complexity and maintenance cost of the system. Secondly, in application scenarios such as transmission lines, electromagnetic radiation will be generated during the operation of power equipment and transmission lines themselves, interfering with the reception of VHF frequency band signals, thereby affecting the detection and positioning of natural lightning by the lightning network monitoring system.

[0042] In recent years, with the development of graphics processing units, the computing power of computers for image processing has been greatly improved, and the ability to use image processing technology to perform industrial monitoring tasks has grown rapidly. Similarly, image detection is also effective in power fields such as transmission line safety assurance. The use of image detection can not only meet the complex and diverse needs of monitoring targets, but also intuitively visualize the data analysis results. At the same time, the background information contained in the image data can also be used for more data analysis, such as environmental assessment, disaster assessment, etc., making the power monitoring system comprehensive and information-based. The transmission line monitoring system based on image detection can extract more structured information and better help maintenance personnel to trace faults and repair damage. However, while the monitoring system obtains richer information, it also faces the pressure of processing and analyzing massive image data, and requires higher real-time and accuracy of image detection algorithms.

[0043] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0044] In a first aspect, an embodiment of the present application provides a method for monitoring lightning in a power transmission line.

[0045] Figure 1 A schematic flow chart of a method for monitoring lightning in a power transmission line in an embodiment of the present application is shown.

[0046] Reference Figure 1 In one embodiment, a method for monitoring lightning in a power transmission line comprises the following steps:

[0047] S1. Monitor the current of the transmission line to determine a target time, wherein the current of the transmission line is greater than or equal to a first preset threshold at the target time and is less than the first preset threshold at a time before the target time.

[0048] In this embodiment, a highly sensitive current sensor is installed on the transmission line to monitor the current changes in the transmission line in real time. Generally, when lightning strikes a transmission line or appears near a transmission line, it will cause a drastic change in the current in the transmission line. The huge energy released by lightning will also affect the current of the transmission line through electromagnetic induction and direct discharge. Therefore, when the current of the transmission line is monitored to suddenly change upward, it means that there is a high probability of lightning.

[0049] S2. Acquire monitoring images of the power transmission line within a period of time before and after the target time as images to be tested.

[0050] In this embodiment, the image acquisition device continuously acquires monitoring images of the transmission line, but only the monitoring images within a period of time before and after the target moment will be used as images to be tested for subsequent image detection, because these monitoring images are most likely to contain monitoring targets, and other monitoring images are considered to be redundant data and will not be subjected to subsequent image detection.

[0051] It is understandable that there is a certain delay between the current mutation phenomenon and the generation of lightning, and the discharge phenomenon of lightning will last for a short time. Therefore, it is necessary not only to obtain monitoring images at the target moment, but also to obtain monitoring images within a certain period of time before and after the target moment.

[0052] Further, in one embodiment, the current of the transmission line is synchronously sampled with the monitoring image of the transmission line, and the sampling frequency is greater than or equal to 10 Hz;

[0053] The image to be tested is the monitoring image of the transmission line within 0.1s before and after the target time.

[0054] In this embodiment, the current of the transmission line is sampled synchronously with the monitoring image of the transmission line to ensure that the monitoring image at the target moment is obtained as the image to be tested. According to experience, there is no monitoring target in the monitoring image with an interval of more than 0.1s from the target moment. The sampling frequency is greater than or equal to 10Hz to ensure that the image to be tested includes at least the monitoring image at the target moment and its previous and next frames of monitoring images.

[0055] S3. Perform image detection on the image to be detected to obtain a determination result of the lightning event.

[0056] Therefore, in this embodiment, the current of the transmission line is monitored to determine the target time, wherein the current of the transmission line is greater than or equal to the first preset threshold at the target time, and is less than the first preset threshold at the moment before the target time; a monitoring image of the transmission line within a period of time before and after the target time is obtained as the image to be tested; and image detection is performed on the image to be tested to obtain the determination result of the lightning event. Through this embodiment, the current impact caused by lightning on the transmission line is utilized to eliminate most of the redundant data without monitoring targets from the monitoring image, thereby reducing the image processing amount, saving the system's equipment computing resources, and reducing the system cost.

[0057] Figure 2 Shows Figure 1 Detailed flowchart of step S3 in FIG.

[0058] Reference Figure 2 , further, in one embodiment, step S3 specifically includes:

[0059] S31, preprocessing the image to be tested to highlight the contour features of lightning;

[0060] S32, inputting the preprocessed image to be tested into a trained target detection model to obtain a target detection result, wherein the target detection model is used to detect lightning objects and power transmission line objects;

[0061] S33. If there are lightning objects and transmission line objects in the target detection result, and the positional relationship between the lightning objects and the transmission line objects meets the preset requirements, the determination result of the lightning event is that there is a risk of lightning striking the transmission line.

[0062] In the event determination method of lightning striking the transmission line, some methods similar to the abnormality determination of the transmission line will mark and train the damaged area and abnormal area of ​​the transmission line image, and use the fitted target detection model to locate the fault area of ​​the transmission line image. This method has the advantages of strong adaptability and high efficiency, but it is not suitable for the determination of lightning events. Due to the instantaneous nature of lightning and the low probability of hitting the line, it is impossible to collect enough training samples in reality. There are more secondary damages such as fire and discharge of the transmission line after the lightning strike, and the training samples cannot achieve consistency with the actual detection targets.

[0063] In this embodiment, in step S32, lightning objects and power transmission line objects are detected by the target detection model, and in step S33, the positional relationship between the lightning objects and the power transmission line objects is used as a basis for determination. Experiments have shown that this method can not only solve the problem of insufficient training samples in traditional methods, but also improve the accuracy of model target detection.

[0064] Transmission lines are mostly located outdoors, where the environment is complex and image detection algorithms are easily affected by background disturbances or changes in lighting. At the same time, outdoor weather such as clouds, rain, fog, and darkness can also affect the visibility and recognition of lightning in monitoring images, making the detection of lightning objects more difficult.

[0065] In this embodiment, in step S31, the contour features of lightning in the image to be detected are highlighted through preprocessing, thereby reducing the difficulty of detecting lightning objects by the target detection model, thereby improving the accuracy of lightning monitoring of power transmission lines.

[0066] Furthermore, in one embodiment, the preset requirements include:

[0067] The center of the detection frame of the lightning object is located in the target area, the center of the target area coincides with the center of the detection frame of the transmission line object, and the ratio of the area of ​​the target area to the area of ​​the detection frame of the transmission line object is a second preset threshold, and the second preset threshold is less than 1.

[0068] Specifically, the detection frame is a rectangle with fixed length and width, and its center position and size are determined according to the specific detection object.

[0069] In this embodiment, whether there is a risk of lightning striking the transmission line is judged based on the center position relationship of the detection frames of the lightning object and the transmission line object. This judgment method is less affected by the length-to-width ratio of the detection frames of the lightning object and the transmission line object and has a high accuracy rate.

[0070] Optionally, the shape of the target area may be a rectangle having the same aspect ratio as the detection frame of the power transmission line object, or may be other shapes.

[0071] By way of example, the second preset threshold is equal to 80%.

[0072] Furthermore, in one embodiment, the preset requirements include:

[0073] The intersection-and-joint ratio of the detection frame of the lightning object and the detection frame of the power transmission line object is greater than a third preset threshold.

[0074] Specifically, the intersection over union (IOU) is calculated by dividing the intersection area of ​​two detection boxes by the union area of ​​the two boxes. The value of the intersection over union ratio is between 0 and 1, and the larger the value, the higher the degree of overlap.

[0075] In this embodiment, whether there is a risk of lightning hitting the transmission line is judged based on the overlap rate of the detection frames of the lightning object and the transmission line object. If the length-to-width ratios of the detection frames of the two types of objects are similar, the accuracy of this judgment method is high. However, when the length-to-width ratios of the detection frames of the two types of objects are greatly different, for example, the detection frame of the transmission line object is usually a rectangle with a width greater than its length, and the length-to-width ratio of the detection frame of the lightning object is uncertain, when the length of the detection frame of the lightning object is greater than its width, even if the centers of the detection frames of the two types of objects coincide, their intersection-to-union ratio is still less than the third preset threshold value. This judgment method may result in missed judgments.

[0076] Optionally, in step S31, the above two determination methods may be combined. Taking the result of a logical OR operation may reduce missed determinations, and taking the result of a logical AND operation may reduce false determinations.

[0077] Figure 3 Shows Figure 2 Detailed flowchart of step S31 in FIG.

[0078] Reference Figure 3 , further, in one embodiment, step S31 specifically includes:

[0079] Convert the image to be tested from RGB color space to Lab color space;

[0080] In Lab color space, contrast-limited adaptive histogram equalization is performed on the brightness channel L;

[0081] Convert the processed image to be tested from Lab color space to RGB color space.

[0082] CLAHE (Contrast Limited Adaptive Histogram Equalization) is an optimization based on AHE (Adaptive Histogram Equalization). AHE is a method to enhance image contrast by adjusting the local contrast of the image to improve the visual effect of the image. CLAHE limits the contrast during equalization to avoid excessive enhancement of noise. The specific steps of CLAHE are as follows:

[0083] 1. Local processing: Divide the image into multiple small area sub-images, and then perform histogram equalization on each sub-image.

[0084] 2. Contrast limitation: Since histogram equalization will over-enhance the noise when processing noisy images, CLAHE will evenly distribute the part of the histogram that exceeds a specific contrast limit to other gray levels after cropping each sub-image to prevent certain gray levels from occupying too many pixels.

[0085] 3. Interpolation merging: To avoid obvious boundaries when merging equalized sub-images, bilinear interpolation is usually used to smooth the transition area to form the final image.

[0086] CLAHE is often only applicable to single-channel images. When facing RGB images, it is necessary to perform histogram equalization on the three channels separately. Although this method is simple and direct, it will cause the color of the processed image to be distorted and unnatural in actual use.

[0087] The RGB color space is an additive color model based on the three primary colors of red, green, and blue. Various colors can be produced by mixing red, green, and blue of different intensities. The Lab color space is a color model based on brightness (L), a channel, and b channel. L represents brightness, a channel represents color components from green to red, and b channel represents color components from blue to yellow. The color range of the Lab color space is wider than that of the RGB color space and can represent more colors.

[0088] In this embodiment, the image to be tested is first converted from the RGB color space to the Lab color space, and then CLAHE is performed only on the brightness channel L, and finally converted back to the RGB color space, so as to enhance the image contrast while maintaining the naturalness of the color as much as possible.

[0089] Furthermore, in one embodiment, the training set used by the target detection model includes original training data and expanded training data generated by the original training data through image data enhancement operations, and the image data enhancement operations include color transformation, scaling, and cropping and splicing;

[0090] The expanded training data generated by color transformation is used to reduce the background interference on the detection of lightning objects and power transmission line objects;

[0091] The expanded training data generated by scaling is used to enhance the adaptability of the target detection model to lightning objects and power transmission line objects of different scales;

[0092] The expanded training data generated by cropping and splicing is used to enhance the adaptability of the target detection model to lightning objects and power transmission line objects under different background, weather and lighting conditions.

[0093] In order to improve its generalization and anti-interference ability to the background, the general algorithm model will rely on a large number of training set samples during the model training phase in order to extract deeper features of the target object. This brings about the demand for the number and diversity of training samples. Since lightning events are relatively rare, the sample data sets that can be collected are relatively small, and it is necessary to perform image data enhancement operations on the original training data to generate expanded training data.

[0094] Specifically in the application scenarios of this embodiment, the background, target size, weather and lighting conditions can easily affect the detection of lightning objects and power transmission line objects. The expanded training data generated by color conversion, scaling, cropping and splicing operations can significantly improve the detection accuracy of the target detection model.

[0095] Optionally, the image data enhancement operation also includes operations such as rotation, translation, shearing, adding noise, and image mixing.

[0096] Image blending is a data enhancement method that combines two images by linear blending to generate new training samples for expanding the training data set. It can also improve the robustness of the model and reduce model overfitting. The implementation steps are:

[0097] 1. Randomly select two images image_1 and image_2 and their corresponding labels label_1 and label_2 from the original training data.

[0098] 2. Use the mixing parameter λ to linearly mix the two images:

[0099] mixed_image=λ×image_1+(1-λ)×image_2.

[0100] 3. Linearly mix the labels using the mixing parameter λ:

[0101] mixed_label=λ×label_1+(1-λ)×label_2.

[0102] Image scaling can be achieved through the bilinear interpolation method. The approximate value of the pixel to be calculated is obtained by performing linear interpolation in two directions using the values ​​of the four known pixels around the pixel to be calculated. Each pixel in the target image is obtained by weighted averaging the pixels in a certain area of ​​the original image. The general operation steps are as follows:

[0103] 1. First, make it clear what target size the original image should be scaled to. For example, if the original image size is M×N, it should be scaled to m×n.

[0104] 2. Calculate the scaling ratios in the horizontal and vertical directions respectively, and set the scaling ratio in the horizontal direction to r x =M / m, the vertical scaling ratio is r y =N / n.

[0105] 3. For each pixel in the target image, determine its corresponding position in the original image. Assuming that the coordinates of a pixel in the target image are (i, j), then its corresponding coordinates in the original image are (i×r x ,j×r y ), recorded as (x, y). This coordinate is usually not an integer, and bilinear interpolation is needed to determine the pixel value of the point. Find the four nearest integer coordinate points around the corresponding position (x1, y1), (x1, y2), (x2, y1), (x2, y2), and the pixel point values ​​are Q 11 , Q 12 , Q 21 , Q 22 .

[0106] 4. Perform two linear interpolations in the horizontal direction and calculate the two corresponding intermediate values ​​in the horizontal direction:

[0107]

[0108] 5. Perform a linear interpolation in the vertical direction to calculate the pixel value of the target pixel:

[0109]

[0110] Normally, considering the equipment implementation and maintenance costs of the transmission line monitoring system, some algorithms or monitoring functions in the system will be migrated to mobile terminals or edge computing devices. The advantage is that it greatly alleviates the pressure of massive data processing of backend services, but it also causes the computing power and storage resources of front-end devices to limit the algorithm itself.

[0111] Optionally, the target detection model is based on the YOLOv5s lightweight detection neural network, and uses a weight-based pruning scheme and sparse regularization to further reduce the model size. The final weight file can be reduced to about 70% of its original size, which can effectively improve the algorithm reasoning capability of the front-end device.

[0112] Optionally, transfer learning is used when training the object detection model. Based on the YOLOv5s pre-trained model, all convolutional layers are frozen in the first 30 iterations, that is, the gradient update of the weight parameters of all convolutional layers is turned off. After 30 iterations, all parameter updates are turned on to fine-tune the model's feature extraction capabilities for private datasets.

[0113] Optionally, the annealing cosine learning rate scheduling strategy is used during target detection model training to dynamically adjust the learning rate parameters, effectively improving the training efficiency of the model while maintaining good generalization performance.

[0114] Optionally, the object detection model is trained using AMP (Automatic Mixed Precision). By mixing FP32 floating point numbers and FP16 floating point numbers during training, the training speed can be significantly accelerated and the use of video memory can be reduced while maintaining the accuracy of the model.

[0115] In a second aspect, an embodiment of the present application also provides a transmission line lightning monitoring device.

[0116] Figure 4 A schematic diagram of functional modules of a lightning monitoring device for a power transmission line in one embodiment of the present application is shown.

[0117] Reference Figure 4 In one embodiment, the transmission line lightning monitoring device comprises:

[0118] A current detection module 10, configured to monitor the current of the transmission line to determine a target time, wherein the current of the transmission line is greater than or equal to a first preset threshold at the target time and is less than the first preset threshold at a time before the target time;

[0119] An image screening module 20 is used to obtain monitoring images of the power transmission line within a period of time before and after the target time as images to be tested;

[0120] The image detection module 30 is used to perform image detection on the image to be detected to obtain a determination result of a lightning event.

[0121] Furthermore, in one embodiment, the image detection module 30 is used to:

[0122] Pre-process the image to be tested to highlight the contour features of lightning;

[0123] The preprocessed image to be tested is input into the trained target detection model to obtain the target detection result, wherein the target detection model is used to detect lightning objects and power transmission line objects;

[0124] If there are lightning objects and transmission line objects in the target detection results, and the positional relationship between the lightning objects and the transmission line objects meets the preset requirements, the determination result of the lightning event is determined as that there is a risk of lightning striking the transmission line.

[0125] Furthermore, in one embodiment, the preset requirements include:

[0126] The center of the detection frame of the lightning object is located in the target area, the center of the target area coincides with the center of the detection frame of the transmission line object, and the ratio of the area of ​​the target area to the area of ​​the detection frame of the transmission line object is a second preset threshold, and the second preset threshold is less than 1.

[0127] Furthermore, in one embodiment, the preset requirements include:

[0128] The intersection-and-joint ratio of the detection frame of the lightning object and the detection frame of the power transmission line object is greater than a third preset threshold.

[0129] Furthermore, in one embodiment, the image detection module 30 is used to:

[0130] Convert the image to be tested from RGB color space to Lab color space;

[0131] In Lab color space, contrast-limited adaptive histogram equalization is performed on the brightness channel L;

[0132] Convert the processed image to be tested from Lab color space to RGB color space.

[0133] Furthermore, in one embodiment, the training set used by the target detection model includes original training data and expanded training data generated by the original training data through image data enhancement operations, and the image data enhancement operations include color transformation, scaling, and cropping and splicing;

[0134] The expanded training data generated by color transformation is used to reduce the background interference on the detection of lightning objects and power transmission line objects;

[0135] The expanded training data generated by scaling is used to enhance the adaptability of the target detection model to lightning objects and power transmission line objects of different scales;

[0136] The expanded training data generated by cropping and splicing is used to enhance the adaptability of the target detection model to lightning objects and power transmission line objects under different background, weather and lighting conditions.

[0137] Further, in one embodiment, the current of the transmission line is synchronously sampled with the monitoring image of the transmission line, and the sampling frequency is greater than or equal to 10 Hz;

[0138] The image to be tested is the monitoring image of the transmission line within 0.1s before and after the target time.

[0139] Among them, the functional implementation of each module in the above-mentioned transmission line lightning monitoring device corresponds to the various steps in the above-mentioned transmission line lightning monitoring method embodiment, and its functions and implementation processes are no longer repeated here one by one.

[0140] In a third aspect, an embodiment of the present application provides a transmission line lightning monitoring device, which may be a personal computer (PC), a laptop computer, a server, or other device with a data processing function.

[0141] Figure 5 A schematic diagram of the hardware structure of a power transmission line lightning monitoring device involved in an embodiment of the present application is shown.

[0142] Reference Figure 5 In an embodiment of the present application, a power transmission line lightning monitoring device may include a processor, a memory, a communication interface, and a communication bus.

[0143] The communication bus may be of any type and is used to interconnect the processor, the memory, and the communication interface.

[0144] The communication interface includes input / output (I / O) interface, physical interface and logical interface, etc., which are used to realize the interconnection of devices inside the power transmission line lightning monitoring equipment, and the interface used to realize the interconnection between the power transmission line lightning monitoring equipment and other devices (such as other computing devices or user devices). The physical interface can be an Ethernet interface, a fiber optic interface, an ATM interface, etc.; the user device can be a display, a keyboard, etc.

[0145] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0146] The processor may be a general-purpose processor, and the general-purpose processor may call the power transmission line lightning monitoring program stored in the memory, and execute the power transmission line lightning monitoring method provided in the embodiment of the present application. For example, the general-purpose processor may be a central processing unit (CPU). The method executed when the power transmission line lightning monitoring program is called may refer to the various embodiments of the power transmission line lightning monitoring method of the present application, and will not be repeated here.

[0147] Those skilled in the art will understand that Figure 5The hardware structure shown in the figure does not constitute a limitation on the present application, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0148] In a fourth aspect, an embodiment of the present application also provides a readable storage medium.

[0149] The readable storage medium of the present application stores a power transmission line lightning monitoring program, wherein when the power transmission line lightning monitoring program is executed by a processor, the steps of the above-mentioned power transmission line lightning monitoring method are implemented.

[0150] Among them, the method implemented when the transmission line lightning monitoring program is executed can refer to the various embodiments of the transmission line lightning monitoring method of the present application, and will not be repeated here.

[0151] It should be noted that the serial numbers of the above-mentioned embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.

[0152] The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The terms "first", "second" and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit "first", "second" and "third" to different types.

[0153] In the description of the embodiments of the present application, "exemplary", "for example" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary", "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "for example" or "for example" is intended to present related concepts in a specific way.

[0154] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or, for example, A / B can mean A or B; the “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two.

[0155] In some processes described in the embodiments of the present application, multiple operations or steps that appear in a specific order are included, but it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or in parallel, and the sequence number of the operation is only used to distinguish the different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in sequence or in parallel, and these operations or steps may be combined.

[0156] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present application.

[0157] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for monitoring lightning in a power transmission line, characterized in that: The transmission line lightning monitoring method comprises: Monitoring the current of the transmission line to determine a target time, wherein the current of the transmission line is greater than or equal to a first preset threshold at the target time and is less than the first preset threshold at a time before the target time; Acquire monitoring images of the power transmission line within a period of time before and after the target time as images to be tested; Image detection is performed on the image to be tested to obtain the determination result of the lightning event.

2. The method for monitoring lightning in a power transmission line according to claim 1, characterized in that: The step of performing image detection on the image to be detected to obtain a determination result of a lightning event comprises: Pre-process the image to be tested to highlight the contour features of lightning; The preprocessed image to be tested is input into the trained target detection model to obtain the target detection result, wherein the target detection model is used to detect lightning objects and power transmission line objects; If there are lightning objects and transmission line objects in the target detection results, and the positional relationship between the lightning objects and the transmission line objects meets the preset requirements, the determination result of the lightning event is determined as that there is a risk of lightning striking the transmission line.

3. The method for monitoring lightning in a power transmission line according to claim 2, characterized in that: Pre-set requirements include: The center of the detection frame of the lightning object is located in the target area, the center of the target area coincides with the center of the detection frame of the transmission line object, and the ratio of the area of ​​the target area to the area of ​​the detection frame of the transmission line object is a second preset threshold, and the second preset threshold is less than 1.

4. The method for monitoring lightning in a power transmission line according to claim 2, characterized in that: Pre-set requirements include: The intersection-and-joint ratio of the detection frame of the lightning object and the detection frame of the power transmission line object is greater than a third preset threshold.

5. The method for monitoring lightning in a power transmission line according to claim 2, characterized in that: The step of preprocessing the image to be tested to highlight the contour features of lightning comprises: Convert the image to be tested from RGB color space to Lab color space; In Lab color space, contrast-limited adaptive histogram equalization is performed on the brightness channel L; Convert the processed image to be tested from Lab color space to RGB color space.

6. The method for monitoring lightning in a power transmission line according to claim 2, characterized in that: The training set used by the target detection model includes the original training data and the expanded training data generated by the original training data through image data enhancement operations. The image data enhancement operations include color transformation, scaling, and cropping and splicing. The expanded training data generated by color transformation is used to reduce the background interference on the detection of lightning objects and power transmission line objects; The expanded training data generated by scaling is used to enhance the adaptability of the target detection model to lightning objects and power transmission line objects of different scales; The expanded training data generated by cropping and splicing is used to enhance the adaptability of the target detection model to lightning objects and power transmission line objects under different background, weather and lighting conditions.

7. The method for monitoring lightning in a power transmission line according to any one of claims 1 to 6, characterized in that: The current of the transmission line is sampled synchronously with the monitoring image of the transmission line, and the sampling frequency is greater than or equal to 10Hz; The image to be tested is the monitoring image of the transmission line within 0.1s before and after the target time.

8. A lightning monitoring device for a power transmission line, characterized in that: The transmission line lightning monitoring device comprises: A current detection module, used to monitor the current of the transmission line to determine the target time, wherein the current of the transmission line is greater than or equal to a first preset threshold at the target time and is less than the first preset threshold at a time before the target time; An image screening module is used to obtain monitoring images of the power transmission line within a period of time before and after the target time as images to be tested; The image detection module is used to perform image detection on the image to be detected to obtain the determination result of the lightning event.

9. A lightning monitoring device for a power transmission line, characterized in that: The power transmission line lightning monitoring device comprises a processor, a memory, and a power transmission line lightning monitoring program stored in the memory and executable by the processor, wherein when the power transmission line lightning monitoring program is executed by the processor, the steps of the power transmission line lightning monitoring method as described in any one of claims 1 to 7 are implemented.

10. A readable storage medium, characterized in that: The readable storage medium stores a power transmission line lightning monitoring program, wherein when the power transmission line lightning monitoring program is executed by a processor, the steps of the power transmission line lightning monitoring method according to any one of claims 1 to 7 are implemented.

Images